Discharge Coefficient for a 1/4" hose 3

[IntrepidLearner]

I'm trying to determine the flow rate for compressed air escaping to the atmosphere via an orifice of known diameter.

Through my search of the web, I managed to come across this link...

http://www.air-dispersion.com/msource.html

It's exactly what I was looking for. With one exception. It suggested an approximate value of 0.72 for the coefficient of discharge (C). However, I've come accross other sources which suggest values anywhere b/w 0.92 and 0.98 for well rounded orifices, down to 0.61 for sharp edges.

I was hoping some gifted individual may actually know what value of C I should use for air escaping through a 1/4" air hose.

 

[BigInch]

Show us your edges.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[IntrepidLearner]

This is the 1st thing that came up when I performed a Google search:

http://www.globalspec.com/FeaturedP...nstrument/Hose_Tubing/12471/0?fromSpotlight=1

 

[Latexman]

I would not treat a 1/4" hose as an orifice, especially if it is long compared to diameter (L/D). A hose is a distributed resistance (antenna) and an orifice is a lumped resistance (resistor). Treat it like pipe and start looking into that.

Good luck,
Latexman

 

[BigInch]

I thought he was talking about entrance/exit losses.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

Latexman (Chemical)I agree with you, however, if the hose is short (or has small losses) then it can be treated as an orifice.

Regards

 

[IntrepidLearner]

I have approx 15 hoses, all approx 1m in length. The exit pressure is 8psi for 10 of them, and 2psi for the others. Should I be considering them as short hoses and/or small losses?

 

[sailoday28]

What is the source (stagnation)pressure and temperature?
I assume the dischare is to atmospheric pressure.

Regards

 

[IntrepidLearner]

I have 100psig in a 1/2" pipe which gets regulated down to 8psig. This pipe then has ten (10) 1/4" offshoot hoses exiting. The discharge is to atmosphere, yes.

Same scenario for the hoses at 2psig. The source temperatures are unknown.

 

[sailoday28]

Assuming the lines are well insulated, you should use a FANNO line approach.
1) A reasonable estimate of the stagnation temperatue must be made.
2) Based on the source pressure, choking is occuring at the control valves. If flow is known, then the equivalent critical flow area can be calculated. If not, then the ideal flow area * Cd must be known.
3) The inlet to the downsteam piping conditions are then determined for 1 and 2 above. With static pressure and Mach number for inlet, I would use an fl/d for smooth piping for the discharge piping.
4) Determine outlet conditions from all of above.

Regards

 

[Latexman]

What size pipe is after the regulator? 1/2"?
What is the line length from the regulator to the first hose?
What is the line length from the first hose to the last hose?

Good luck,
Latexman

 

[IntrepidLearner]

OK, sailoday, I'm confused on a few fronts.

1st, you lost me when you started talking about mach number, fl/d, critical flow area. I'm 'only' an electrical engineer .

2nd, I'm not sure where you've got these control valves coming from. These air hoses are open to atmosphere. Each is fed into an electrical panel with vents on it, with the aim of creating a higher pressure inside than outside. The reasoning behind this was to keep our elec panels, which are situated in a very dusty environment, dust-free. Whether it's working and is energy efficient is another story.

3rd, I believe that in both situations, the flow is non-choked since the pressure in the hoses (which I had assumed was the source pressure) is less than 1.7-1.9 that of the destination pressure (atmosphere).

 

[IntrepidLearner]

I'll check on those things for you, Latexman...

 

[Compositepro]

You might want to rethink your project. You will probably be spending many thousands of dollars per year in electricity using compressed air as you describe. You will be far better-off using a small 1/2 Hp blower and 1" PVC pipe to distribute air to your enclosures. You get less than eight scfm of air per 1 Hp if you use 100 psi air as your source. Or put a fan and filter on each cabinet. But your filters would probably plug quickly.

Air is not free, as we would like to think.

 

[MikeHalloran]

Several comments:

I have done some work with orifices. When I need a predictable drop, I specify sharp edges, no burr, no radius, no chamfer, and use an orifice coefficient of 0.61. I have had occasional problems with craftsmen who couldn't leave them sharp, or who radiused them "so they'd flow better". They do, but not predictably. The coefficient is extremely sensitive to the edge radius.

Hose is usually connected to pipe with barbed adapters. For 1/4" hose, the ID of a typical metal adapter is going to be 3/16", maybe a little less. You need to account for the ID of the actual adapters you intend to use.

I hope you're using clean dry instrument air for purging. The typical factory's compressed air supply is wet enough to cause you a load of headaches in an electrical box.

Is there some reason you can't use sealed boxes?







Mike Halloran
Pembroke Pines, FL, USA

 

[IntrepidLearner]

Latexman:

For hoses discharging at 8psi:
[a] size before and after regulator is 1/2"
length from regulator to 1st hose = 10'
[c] to last hose = 10' + 15.25' = 25.25'
- Number of hoses is actually 14
- Length of each hose is approx 2'
- 1/2" pipe from regulator braches into two upon entry into elec room. Each leg having 7 hoses along it's length. And each (pretty much) a mirror of the other.

For hoses discharging at 2psi:
[a] size before and after regulator is 1/2"
length from regulator to 1st hose = 29.25'
[c] to last hose = 29.25' + 15.25' = 44.5'
- Number of hoses is actually 4
- Length of each hose is approx 2'

 

[IntrepidLearner]

To Composite & MikeH:

How I've described the air distribution to our elec panels is actually in place now - and has been so for years. My goal is to show them how much compressed air/energy/$ they're wasting, and to suggest a possible solution. Furthermore, the whole premise of creating a pressure differential between the inside and outside of these panels comes apart when someone leaves any of the elec panel doors open - which often happens!


To MikeH:

So you're saying that the formula I quoted in my original post is applicable? And that I should be using a Cd of 0.61?

 

[MikeHalloran]

No, it's not applicable here.

WHEN you use orifices, at some other time, and you want them predictable, use sharp edges and a coefficient of 0.61.



Mike Halloran
Pembroke Pines, FL, USA

 

[katmar]

Hi IntrepidLearner,

The problem you are dealing with here is not trivial unless you have some good software for flow of compressible fluids in networks, which I doubt an Electrical man would have in his toolbag. However, for you to be able to do an either/or analysis between piped air and your alternative solution a simple assumption makes a spreadsheet or hand calculation possible.

The reason I say it is not trivial is that you will lose a significant portion of your available pressure due to friction pressure drop between the regulator and the first 1/4" nozzle. Additionally, you will continue to lose pressure due to friction as the air flows down the pipe past all the nozzles, so that each nozzle has a lower supply pressure than the one before it. These two factors mean that you have to do a trial and error calculation, until you get a converged result.

When process engineers design manifolds like this we often "cheat". If the pressure drop down the manifold is small compared with the pressure drop through the branches (or orifices or nozzles in similar situations) you can regard the pressure in the manifold as constant at its supply pressure all along its length and then every branch calculation is identical and the problem does become trivial. To take your example of 7 off 1/4" branches on a 1/2" pipe - the first bit of pipe has 7 times the flow that a branch has, but it has only 4 times the area. This makes the pressure drop in the pipe relatively significant. As a rule-of-thumb the pressure drop decreases with the 5th power of the diameter, so a 1" pipe would have a pressure drop of 1/32 of that in the 1/2" pipe. If your manifold was 1" you could make these assumptions. Although I called this a "cheat" it does result in every branch having almost exactly the same flowrate, and this can have a process benefit.

Anyway, this is probably far more detail than you need or are interested in! Let's get to how you can actually do the calculation.

The assumption that I would propose to make the problem easily soluble in your case is to assume that the flow through each branch is identical. You really need to test this, but because the branches are relatively close together and because the flowrate (and therefore the pressure drop per foot) drops rapidly as you go along the manifold, it is reasonable in your case. Although you can probably ignore this pressure drop along the manifold between the first and last branches, you cannot ignore the pressure drop between the regulator and the first nozzle (i.e. you cannot assume each branch sees the full supply pressure)

You are now faced with a trial and error calculation to determine the pressure at the first branch. This can be done by

1. Guess a total flow from the regulator to the first branch
2. Based on this flow, calculate the pressure drop in the manifold from the regulator to the point of the first branch
3. From this pressure drop, you know the supply pressure to the branch (assumed the same for all n branches)
4. Calculate the flow per branch and multiply by the number of branches to get total flow
5. If the result from step 4 is different from the flow you used in step 2, take the average of these two and go back to step 2.

In determining the pressure drop in the branches I would take each branch as
one inlet loss,
one "thick orifice" for the bayonet connector
2' of hose
one exit loss

If you have a problem with these concepts ask your chemical or mechanical engineering colleagues for a copy of Crane Technical Publication #410. I suppose any fluids book would be OK, but the Crane book is very practically oriented.

I have had to guess some of your dimensions and temperatures, but my estimates are that in the 8 psi system you will get about 5.0 scfm per branch, and in the 2 psi system you will get about 2.6 scfm per branch. Standard conditions taken as 60 deg F and 14.7 psia.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[Latexman]

IntrepidLearner,

katmar has summed the situation up nicely. It's an extremely tough problem for a beginner at fluid flow.

katmar - I think you may have missed the following point:

1/2" pipe from regulator braches into two upon entry into elec room. Each leg having 7 hoses along it's length. And each (pretty much) a mirror of the other.

Which makes it even more complicated if the distance from the regulator to the tee is significant.

My recommendation is to farm this one out to someone that has the knowledge and software to do it. Is that an option?



Good luck,
Latexman